Part I - Ford GoBike System Dataset

by Saoban Lateefat

Introduction

This dataset is based on individual rides madein a bike-sharing system covering the greater San FranciscoBay area.

Preliminary Wrangling

# import all packages and set plots to be embedded inline
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import plotly.express as px

%matplotlib inline

bike = pd.read_csv('201902-fordgobike-tripdata.csv')

print(bike.shape)
print( )
bike.info()
print( )
bike.head()

(183412, 16)

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 183412 entries, 0 to 183411
Data columns (total 16 columns):
 #   Column                   Non-Null Count   Dtype  
---  ------                   --------------   -----  
 0   duration_sec             183412 non-null  int64  
 1   start_time               183412 non-null  object 
 2   end_time                 183412 non-null  object 
 3   start_station_id         183215 non-null  float64
 4   start_station_name       183215 non-null  object 
 5   start_station_latitude   183412 non-null  float64
 6   start_station_longitude  183412 non-null  float64
 7   end_station_id           183215 non-null  float64
 8   end_station_name         183215 non-null  object 
 9   end_station_latitude     183412 non-null  float64
 10  end_station_longitude    183412 non-null  float64
 11  bike_id                  183412 non-null  int64  
 12  user_type                183412 non-null  object 
 13  member_birth_year        175147 non-null  float64
 14  member_gender            175147 non-null  object 
 15  bike_share_for_all_trip  183412 non-null  object 
dtypes: float64(7), int64(2), object(7)
memory usage: 22.4+ MB

	duration_sec	start_time	end_time	start_station_id	start_station_name	start_station_latitude	start_station_longitude	end_station_id	end_station_name	end_station_latitude	end_station_longitude	bike_id	user_type	member_birth_year	member_gender	bike_share_for_all_trip
0	52185	2019-02-28 17:32:10.1450	2019-03-01 08:01:55.9750	21.0	Montgomery St BART Station (Market St at 2nd St)	37.789625	-122.400811	13.0	Commercial St at Montgomery St	37.794231	-122.402923	4902	Customer	1984.0	Male	No
1	42521	2019-02-28 18:53:21.7890	2019-03-01 06:42:03.0560	23.0	The Embarcadero at Steuart St	37.791464	-122.391034	81.0	Berry St at 4th St	37.775880	-122.393170	2535	Customer	NaN	NaN	No
2	61854	2019-02-28 12:13:13.2180	2019-03-01 05:24:08.1460	86.0	Market St at Dolores St	37.769305	-122.426826	3.0	Powell St BART Station (Market St at 4th St)	37.786375	-122.404904	5905	Customer	1972.0	Male	No
3	36490	2019-02-28 17:54:26.0100	2019-03-01 04:02:36.8420	375.0	Grove St at Masonic Ave	37.774836	-122.446546	70.0	Central Ave at Fell St	37.773311	-122.444293	6638	Subscriber	1989.0	Other	No
4	1585	2019-02-28 23:54:18.5490	2019-03-01 00:20:44.0740	7.0	Frank H Ogawa Plaza	37.804562	-122.271738	222.0	10th Ave at E 15th St	37.792714	-122.248780	4898	Subscriber	1974.0	Male	Yes

bike.columns

Index(['duration_sec', 'start_time', 'end_time', 'start_station_id',
       'start_station_name', 'start_station_latitude',
       'start_station_longitude', 'end_station_id', 'end_station_name',
       'end_station_latitude', 'end_station_longitude', 'bike_id', 'user_type',
       'member_birth_year', 'member_gender', 'bike_share_for_all_trip'],
      dtype='object')

bike.describe()

	duration_sec	start_station_id	start_station_latitude	start_station_longitude	end_station_id	end_station_latitude	end_station_longitude	bike_id	member_birth_year
count	183412.000000	183215.000000	183412.000000	183412.000000	183215.000000	183412.000000	183412.000000	183412.000000	175147.000000
mean	726.078435	138.590427	37.771223	-122.352664	136.249123	37.771427	-122.352250	4472.906375	1984.806437
std	1794.389780	111.778864	0.099581	0.117097	111.515131	0.099490	0.116673	1664.383394	10.116689
min	61.000000	3.000000	37.317298	-122.453704	3.000000	37.317298	-122.453704	11.000000	1878.000000
25%	325.000000	47.000000	37.770083	-122.412408	44.000000	37.770407	-122.411726	3777.000000	1980.000000
50%	514.000000	104.000000	37.780760	-122.398285	100.000000	37.781010	-122.398279	4958.000000	1987.000000
75%	796.000000	239.000000	37.797280	-122.286533	235.000000	37.797320	-122.288045	5502.000000	1992.000000
max	85444.000000	398.000000	37.880222	-121.874119	398.000000	37.880222	-121.874119	6645.000000	2001.000000

bike.isnull().sum()

duration_sec                  0
start_time                    0
end_time                      0
start_station_id            197
start_station_name          197
start_station_latitude        0
start_station_longitude       0
end_station_id              197
end_station_name            197
end_station_latitude          0
end_station_longitude         0
bike_id                       0
user_type                     0
member_birth_year          8265
member_gender              8265
bike_share_for_all_trip       0
dtype: int64

bike['member_birth_year'].fillna(bike['member_birth_year'].mode()[0], inplace = True)
bike['member_gender'].fillna(bike['member_gender'].mode()[0], inplace = True)
bike.dropna(inplace = True)
bike.isnull().sum()

duration_sec               0
start_time                 0
end_time                   0
start_station_id           0
start_station_name         0
start_station_latitude     0
start_station_longitude    0
end_station_id             0
end_station_name           0
end_station_latitude       0
end_station_longitude      0
bike_id                    0
user_type                  0
member_birth_year          0
member_gender              0
bike_share_for_all_trip    0
dtype: int64

bike['start_time'] = pd.to_datetime(bike['start_time'], format= '%Y-%m-%d %H:%M:%S')
bike['end_time'] = pd.to_datetime(bike['end_time'], format= '%Y-%m-%d %H:%M:%S')

bike['member_birth_year'] = bike['member_birth_year'].astype(int)

bike.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 183215 entries, 0 to 183411
Data columns (total 16 columns):
 #   Column                   Non-Null Count   Dtype         
---  ------                   --------------   -----         
 0   duration_sec             183215 non-null  int64         
 1   start_time               183215 non-null  datetime64[ns]
 2   end_time                 183215 non-null  datetime64[ns]
 3   start_station_id         183215 non-null  float64       
 4   start_station_name       183215 non-null  object        
 5   start_station_latitude   183215 non-null  float64       
 6   start_station_longitude  183215 non-null  float64       
 7   end_station_id           183215 non-null  float64       
 8   end_station_name         183215 non-null  object        
 9   end_station_latitude     183215 non-null  float64       
 10  end_station_longitude    183215 non-null  float64       
 11  bike_id                  183215 non-null  int64         
 12  user_type                183215 non-null  object        
 13  member_birth_year        183215 non-null  int32         
 14  member_gender            183215 non-null  object        
 15  bike_share_for_all_trip  183215 non-null  object        
dtypes: datetime64[ns](2), float64(6), int32(1), int64(2), object(5)
memory usage: 23.1+ MB

bike_clean = bike.copy()

bike_clean['start_day'] = bike_clean['start_time'].dt.day
bike_clean['start_month'] = bike_clean['start_time'].dt.month
bike_clean['start_year'] = bike_clean['start_time'].dt.year
bike_clean['start_weekday'] = bike_clean['start_time'].dt.weekday
bike_clean['start_weekday_name'] = bike_clean['start_time'].dt.day_name()

bike_clean['end_day'] = bike_clean['end_time'].dt.day
bike_clean['end_month'] = bike_clean['end_time'].dt.month
bike_clean['end_year'] = bike_clean['end_time'].dt.year
bike_clean['end_weekday'] = bike_clean['end_time'].dt.weekday
bike_clean['end_weekday_name'] = bike_clean['end_time'].dt.day_name()

bike_clean['duration_min'] = round(bike_clean['duration_sec']/60)
bike_clean['duration_min']

0          870.0
1          709.0
2         1031.0
3          608.0
4           26.0
           ...  
183407       8.0
183408       5.0
183409       2.0
183410       2.0
183411       5.0
Name: duration_min, Length: 183215, dtype: float64

bike_clean['member_age'] = (2019 - bike_clean['member_birth_year'])
bike_clean['member_age']

0         35
1         31
2         47
3         30
4         45
          ..
183407    23
183408    35
183409    29
183410    31
183411    30
Name: member_age, Length: 183215, dtype: int32

#bike_clean.shape
bike_clean.columns

Index(['duration_sec', 'start_time', 'end_time', 'start_station_id',
       'start_station_name', 'start_station_latitude',
       'start_station_longitude', 'end_station_id', 'end_station_name',
       'end_station_latitude', 'end_station_longitude', 'bike_id', 'user_type',
       'member_birth_year', 'member_gender', 'bike_share_for_all_trip',
       'start_day', 'start_month', 'start_year', 'start_weekday',
       'start_weekday_name', 'end_day', 'end_month', 'end_year', 'end_weekday',
       'end_weekday_name', 'duration_min', 'member_age'],
      dtype='object')

What is the structure of your dataset?

The dataset originally has 183412 rows and 16 columns which are 'duration_sec', 'start_time', 'end_time', 'start_station_id', 'start_station_name', 'start_station_latitude', 'start_station_longitude', 'end_station_id', 'end_station_name', 'end_station_latitude', 'end_station_longitude', 'bike_id', 'user_type'user typr shou, 'member_birth_year', 'member_gender', 'bike_share_for_all_trip' but after wrangling the dataset, we have 183215 rows and 28 columns which are duration_sec', 'start_time', 'end_time', 'start_station_id', 'start_station_name', 'start_station_latitude', 'start_station_longitude', 'end_station_id', 'end_station_name', 'end_station_latitude', 'end_station_longitude', 'bike_id', 'user_type', 'member_birth_year', 'member_gender', 'bike_share_for_all_trip', 'start_day', 'start_month', 'start_year', 'start_weekday', 'start_weekday_name', 'end_day', 'end_month', 'end_year', 'end_weekday', 'end_weekday_name', 'duration_min'and 'member_age'.

What is/are the main feature(s) of interest in your dataset?

I did be looking out for insights on how the columns affects the bike system

What features in the dataset do you think will help support your investigation into your feature(s) of interest?

I think weekday name, member age , user type should really help me in my investigation.

Univariate Exploration

I start by looking at the gender

Count per gender

base_color = sns.color_palette()[0]
plt.title("Count per gender ")
gender = bike_clean['member_gender'].value_counts().index
sns.countplot(data=bike_clean, y='member_gender', color=base_color, order=gender);

The male gender has rides bikes the most with a count of 138763 and least with the others(i.e gender not identified)

user type per count

count = bike_clean['user_type'].value_counts()
label = ['Subscriber', 'Customer']
colors = ["#8c564b", "#ff7f0e"]
fig = plt.figure(figsize =(10, 7))
plt.pie(count, labels = label, autopct='%1.1f%%', colors=colors)
plt.title("User Type ")
 
# show plot
plt.show()

We have subscribers of 89.2% and Customers of 10.8% Most people who rides ford gobikes are more of subcribers

Checking for the percentage of people who do share trips and those who do not

count = bike_clean['bike_share_for_all_trip'].value_counts()
label = ['No', 'Yes']
colors = ["#8c564b", "#ff7f0e"]
fig = plt.figure(figsize =(10, 7))
plt.pie(count, labels = label, autopct='%1.1f%%', colors=colors)
plt.title("bike share for all trip ")
 
# show plot
plt.show()

Here, we can see we have more of people who do not share rides with a total of 90.5% to people who do share rides which is just 9.5%

Duration of rides per seconds

plt.figure(figsize = (6, 4), dpi = 100)

sns.displot(bike_clean, x="duration_sec");#, kde = False)
plt.xlim(0, 4000)
plt.title("Trip per seconds'")
plt.xlabel('Duration in secons')
plt.ylabel('Frequency')
plt.show();

<Figure size 600x400 with 0 Axes>

fig = px.histogram(bike_clean, x = 'duration_min',  title = 'Rides per minutes')
fig.show()

average = sum(bike_clean['duration_min'])/len(bike_clean['duration_min'])
round(average, 0)

12.0

The duration with the highest rides are those with the short minutes and we can see that most rised occurs for just 6 minutes and also the average minutes for each ride is 12 minutes.

Age range for each rides

fig = px.histogram(bike_clean, x = 'member_age', title = 'Age per rides')
fig.show()

Taking further insights by looking at the ages of people who ride Ford GoBikes the most, We can see most of the people who rides bikes are between the age of 22 and 40 which are mostly work age range.

Weekday Per Rides

weekdays = ['Sunday', 'Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday']
fig, ax = plt.subplots(nrows=2, figsize = [8,8])

#plt.title = 'Rides per weekdays'
default_color = sns.color_palette()[0]
sns.countplot(data = bike_clean, x = 'start_weekday_name', color = default_color, ax = ax[0], order = weekdays).set(title = 'Rides per weekdays')
sns.countplot(data = bike_clean, x = 'end_weekday_name', color = default_color, order = weekdays)

plt.show()

Here, we can see most rides occurs on weekdays and mostly on Thursdays to weekends, we can say this is due prolly because they are workdays.

Trip per hours

bike_clean['tim'] = bike_clean['start_time'].dt.time
bike_clean['hh'] = bike_clean['start_time'].dt.strftime('%H')

plt.figure(figsize = [15, 10]);
base_color = sns.color_palette()[0]
plt.title("Count per hour ")
hour = bike_clean['hh'].value_counts().index
sns.countplot(data=bike_clean, y='hh', color = base_color, order=hour);

Most trips occurs around 5pm and 8am, and its very obvious this is due to rush hours i.e the time most people do leave work and time most people do leave for work.

Top 10 Stations

fig, ax = plt.subplots(nrows=2, figsize = [8,8])
bike_clean['start_station_name'].value_counts()[:10].plot(kind='barh', ax = ax[0], title = 'Top 10 start_station')
bike_clean['end_station_name'].value_counts()[:10].plot(kind='barh', title = 'Top 10 end_station');

Noticed we have similar stations for the top 10 start and end stations

Discuss the distribution(s) of your variable(s) of interest. Were there any unusual points? Did you need to perform any transformations?

The male gender has the most records,Most of people who takes the trips are subscribers, Ages that takes rides the most is age 31 and most trips ocuurs for just 6 mins.

Of the features you investigated, were there any unusual distributions? Did you perform any operations on the data to tidy, adjust, or change the form of the data? If so, why did you do this?

No there isn't any unusual distribution in the dataset.

Bivariate Exploration

Gender per user type

sns.countplot(data = bike_clean, x = 'member_gender', hue = 'user_type').set(title = 'gender per user types');

We have more males who are subscribers to female

Gender per bike share

sns.countplot(data = bike_clean, x = 'member_gender', hue = 'bike_share_for_all_trip').set(title = 'gender per bike share');

Most males do not share trips

Trip duration per days

base_color = sns.color_palette()[0]
count = bike_clean['start_weekday_name'].value_counts().index
plt.title("Count per weekday ")
gender = bike_clean['start_weekday_name'].value_counts().index
sns.lineplot(data=bike_clean, x='start_weekday_name', y='duration_min', color=base_color);
plt.xticks(rotation=90);

Trips are faster on weekdays to weekends

Talk about some of the relationships you observed in this part of the investigation. How did the feature(s) of interest vary with other features in the dataset?

In all, We have more males who are subscribers to female, Most males do not share trips and Trips are faster on weekdays to weekends.

Multivariate Exploration

ax = sns.barplot(data = bike_clean, x = 'member_gender', y = 'duration_min', hue = 'user_type')
#plt.title("Relationship between gender, duration in minutes and user type ")
ax.legend(loc = 8, ncol = 3, framealpha = 1, title = 'user type');

-- Customers spends more time on rides than the subscribers
-- The male and female gender spends averagely the same time on rides

ax = sns.pointplot(data = bike_clean, x = 'member_gender', y = 'duration_min', hue = 'bike_share_for_all_trip', dodge = 0.3, linestyles = "").set(title = "Relationship between gender, duration in minutes and bike_share_for_all_trip  ")

The male gender spends less time and also have the least of those who share trips

plt.figure(figsize = [15, 10]);
sns.scatterplot(data=bike_clean, x="member_age", y="duration_min", hue="member_gender", style="member_gender").set(title = "Relationship between member age, duration in minutes and gender");

Younger males tends to spends more time on rides to older males

plt.figure(figsize = [15, 10]);
sns.scatterplot(data=bike_clean, x="member_age", y="duration_min", hue="user_type", style="member_gender").set(title = "Relationship between member age, duration in minutes and gender");

Talk about some of the relationships you observed in this part of the investigation. Were there features that strengthened each other in terms of looking at your feature(s) of interest?

We can see Customers spends more time on rides than the subscribers, and also most people whose gender is not identified spends more time to those whose gender is identified, So, we can say the time spent on the rides doesn't really depends on whether the person is a male or a female sine they spends equal minutes on rides.

Were there any interesting or surprising interactions between features?

The male gender are the least of those who share trips

Conclusions

After the wrangling and visualizations, We can infer that:
-- We have more males to female
-- There are more of people who are subscribers to customers
-- Most trips are done within short hours.
-- We have higher trips on weekdys to weekends
-- Most ofthe trip is made around 5pm and 8am and we deduce this is due to rush hours.
-- We have younger people who board bikes to older people